|Publication number||US4556910 A|
|Application number||US 06/568,319|
|Publication date||Dec 3, 1985|
|Filing date||Jan 5, 1984|
|Priority date||Jan 5, 1984|
|Publication number||06568319, 568319, US 4556910 A, US 4556910A, US-A-4556910, US4556910 A, US4556910A|
|Original Assignee||Fuji Photo Film Co., Ltd.|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (2), Referenced by (22), Classifications (7), Legal Events (4)|
|External Links: USPTO, USPTO Assignment, Espacenet|
The invention pertains to a solid-state image sensing device including a fixed-pattern noise reducing circuit.
FIG. 1 is a diagram illustrating a prior art image sensing device of the same general type to which the invention pertains. The device includes a sensor array chip 15 upon which are fabricated a pixel (picture element) array 12, vertical and horizontal shift registers 11 and 13, and output switches comprising FETs 19. The pixel array 12 includes vertical lines 18 and horizontal lines 20 arranged perpendicular to one another. (Although the vertical lines 18 extend in the horizontal direction in the figure and the horizontal lines 20 in the vertical direction, nevertheless, they are properly termed vertical and horizontal lines, respectively, since the vertical lines 18 are scanned in the vertical direction and the horizontal lines 20 scanned in the horizontal direction.) Light-sensing pixels 14 are formed at each intersection between a vertical line 18 and horizontal line 20. Each pixel schematically is composed of an FET switch 17 and a photosensitive diode 16.
The horizontal lines 20 are connected to the sources of respective output switch FETs 19, the drains of which are connected in common to a line 37. The gates of the FETs are connected to respective outputs of the horizontal shift register 13. The line 37 is connected to a video output terminal pad 23. The terminal pad 23 is coupled externally of the chip 15 to the inverting input of an operational amplifier 41, the noninverting input of which is connected to a source of a reference potential VREF. Alternatively, the video output can be summed with a reference potential and the sum applied to the inverting input of the amplifier 41. A feedback resistor 42 is connected between the output of the amplifier 41 and the inverting input. The output of the amplifier 41 is also applied to a fixed-pattern noise integrator circuit 43.
In operation, the vertical shift register is operated by clocking signals φV1 and φV2 to activate the lines 18 in sequence. Activation of each of the lines 18 in this manner causes the FETs 17 having gates connected to that line to turn on, thereby transferring signal charges, the magnitudes of which are indicative of the sensed light intensity, from the respective photosensitive diodes 16 to the horizontal lines 20. After the signal charges have been transferred, the horizontal shift register 13 is operated by clocking signals φH1 and φH2 to turn on the FETs 19 in sequence, thereby to transfer the signal charges from the horizontal lines 20 one at a time to the line 37, and thence via the terminal pad 23 to the inverting input of the amplifier 41.
Referring now to the waveform diagrams of FIGS. 2A through 2F, the operation of the abovedescribed device will be further explained. FIGS. 2A and 2B show two pulses from adjacent outputs of the horizontal shift register 13. In response to the pulses of FIGS. 2A and 2B, signal charges, as illustrated by the current signals shown in FIGS. 2C and 2D, from corresponding horizontal lines 20 are transferred via respective FETs 19 to the output line 37. However, the pulses of FIGS. 2C and 2D do not appear in this form on the output line 37. Instead, due to the presence of stray capacitances 31 through 36, the current signal on the output line 37 actually appears as shown in FIG. 2E because of switching transients and induced noise. In fact, as can readily be appreciated from comparing FIGS. 2C and 2D with FIG. 2E, the switching transients and noise components are much greater in magnitude than the actual signal components. The actual signal components show up in the waveform of FIGS. 2E as only a slight drooping in the waveform, as indicated by the difference between the dashed line and adjacent solid-line portions of the indicated waveform. The peak amplitudes of the waveform of FIG. 2E carry little information regarding the signal amplitude. Moreover, the peaks of the waveform FIG. 2E vary from pulse to pulse, thereby resulting in what is known as fixed-pattern noise. The subject of fixed-pattern noise is discussed in more detail in the articles: Nabeyama et al., "All Solid-State Color Camera with Single Chip MOS Imager", IEEE Transactions on Consumer Electronics, Vol. CE-27, February 1981; Imada et al., "Noise of an MOS-type Area Imager", 1982 Conference on Custom Integrated Circuits, Rochester, N.Y.; Aoki et al., "2/3 Inch Format MOS Single Chip Color Imager", IEEE Transactions on Electron Devices, Vol. 29, No. 4, April 1982; and Hodges et al., "Potential of MOS Technologies for Analog Integrated Circuits", IEEE Journal of Solid State Circuits, Vol. SC-13, June 1971, pp. 285-294.
To extract the signal component from the signal of FIG. 2E and to eliminate the fixed-pattern noise, the signal from the output terminal pad 23 is buffered and amplified by the amplifier 41, which converts the current signal to a voltage signal, and then integrated by the fixed-pattern noise integrator 43. As indicated in FIG. 2F, the integrator 43 integrates the amplified signal starting from a time t0 at the rising edge of a pulse from the horizontal shift register 13 and stops the integration at a time t1 after all switching transients have died out. Between the time t1 and a subsequent time t2, the output from the integrator 43 is read by a subsequent processing circuit (not shown), after which the integrator 43 is reset (with a signal RESET) between times t2 and t3 to prepare for the next pulse.
The arrangement of FIG. 1 suffers from the following difficulties: First, the amplifier 41 must have a gain-bandwidth product (GBW) of about 400 to 600 MHz for ordinary video rates to avoid signal distortion which would prevent accurate signal integration. Also, the feedback resistor 42 must have a resistance value in excess of 100 kΩ to produce a usable signal strength. This high resistance value makes the resistor 42 prone to parasitic effects at high frequencies. Still further, the magnitude of the stray capacitance 36 is many times larger than the internal chip capacitance, thereby increasing the random noise. Finally, the large gain-bandwidth product of the amplifier 41 makes the construction of this amplifier unavoidably complex, making it large in size, and further increasing the magnitude of the stray capacitance 36.
To reduce the magnitude of the stray capacitance 36, it is possible to move the amplifier 41, feedback resistor 42 and fixed-pattern noise integrator 43 onto the sensor array chip 15. However, doing so is accompanied by other problems. Specifically, the gain-bandwidth product which can be achieved using current MOS integrated circuit techniques is only about 10 to 20 MHz. Also, it is difficult to fabricate the high resistance value feedback resistor and to make the resistance value sufficiently accurate.
Accordingly, it is an object of the present invention to provide an image sensing device which avoids the specific drawbacks of the prior art approaches mentioned above.
This, as well as other objects of the invention, are met by an image sensing device including an array of photosensing elements, means for sequentially coupling outputs of the photosensing elements to an array output terminal, and an output signal extracting circuit including an operational amplifier having an input connected to the array output terminal and a resettable feedback capacitor connected directly between the input and output of the amplifier. More specifically, the output signal extracting circuit includes an operational amplifier having a noninverting input coupled to a reference potential source and an inverting input coupled to the array output terminal. The feedback capacitor is coupled between the output of the operational amplifier and the inverting input. An FET transistor, provided to reset the feedback capacitor, has a channel coupled in parallel with the feedback capacitor and a gate coupled to a source of a reset signal. A second capacitor is coupled between the inverting input and a source of a signal having a waveshape similar to, but a polarity opposite from, a non-information-bearing component of the signal on the array output terminal. All of the above components are preferably fabricated upon a single semiconductor chip. Further, a buffer amplifier is provided on the chip for isolating the operational amplifier from the output pin and stray capacitance. For the source of the signal having a waveshape similar to but polarity opposite from the non-information-bearing component of the signal on the array output terminal, a source of the reset signal inverted may be employed. Otherwise, a resistive summing network receiving as inputs the the inverted reset signal and inverted clocking signals may be used.
FIG. 1 is a diagram showing a prior art image sensing device of the same general type to which the invention pertains;
FIGS. 2A through 2F are waveform diagrams utilized to explain the operation of the circuitry of FIG. 1;
FIG. 3 is a schematic diagram of an improved output signal extracting circuit of the invention; and
FIG. 4 is a schematic diagram of an alternative embodiment of an output signal extracting circuit of the invention.
In accordance with the invention, the above-noted drawbacks of the prior art image sensing device are obviated by disposing of the operational amplifier 41, feedback resistor 42 and fixed-pattern noise integrator 43 utilized in the circuitry of FIG. 1, and adding, on the sensor array chip, an output signal extracting circuit as shown in FIG. 3. In FIG. 3, the inverting input of an operational amplifier 50 is connected to the output line 37. The noniverting input of the operational amplifier 50 receives the reference potential VREF. A feedback capacitor 52 is connected directly between the output and the inverting input of the amplifier 50. The capacitance value of the capacitor 52 is preferably in a range of 0.3 to 1 pf. An FET 53 is provided having a channel shunting the capacitor 52. The gate of the FET 53 receives the RESET signal. A second capacitor 54, formed by connecting together the source and drain of an FET, is connected between the inverting input of the amplifier 50 and a source of the RESET signal inverted (RESET). In some situations, however, the capacitor 54 and source of the reset signal inverted can be omitted, specifically, if the noise level is sufficiently low. The output of the amplifier 50 is buffered by a buffer amplifier 51, and then applied to the output terminal pad 23. The output terminal pad 23 is connected directly via the output line 44 to the external processing circuitry.
In operation, the scanning of the pixel array 12 by the vertical shift register 11 and the sequential coupling of the outputs from the pixel array 12 to the output line 37 by operation of the horizontal shift register 13 takes place in the same manner as described above. Each time that a pulse is presented to the output signal extracting circuitry of FIG. 3 on the output line 37, the pulse is integrated upon the feedback capacitor 52. The feedback capacitor 52 is reset after each integration period.
It is the purpose of the second capacitor 54 to nominally compensate for the switching transients produced due to the gate-drain capacitance of the FET 53. By feeding to the inverting input of the amplifier 50 the signal RESET, the switching transients produced at the transitions of the RESET signal are substantially cancelled.
If the switching transients on the output line 37 include transients produced, for example, by the clock signals φH1 and φH2 utilized by the horizontal shift register 13, the arrangement of FIG. 4 can be utilized. More specifically, the signals which give rise to the unwanted switching transients are summed in a resistive network composed of resistors 61 through 64 and applied to one side of the capacitor 54 for capacitive coupling to the inverting input of the amplifier 50. Although the signals RESET, φH1 and φH2 are indicated in the embodiment of FIG. 4, any signal which gives rise to an unwanted transient component can be utilized. The values of the resistors of the network should, of course, be adjusted in accordance with the relative magnitudes of the various transient components.
The invention provides the following advantages: The stray capacitance on the output terminal of the image sensing device is drastically reduced. Secondly, the integrator amplifier need not have such a high gain-bandwidth product because the integration function is performed directly. Hence, the amplifiers 50 and 51 can be readily implemented with standard integrated circuit techniques. Their gain-bandwidth product need only be so high as to allow settling to occur before the next reset pulse. Also, the amplifier 50 may actually clip or distort the signal without causing cumulative errors. No active external amplifier connection is required to the amplifier 50, thereby permitting the use of a small integrating capacitor and simple amplifier design. Finally, the gain-bandwidth product of the output buffer amplifier 51 need only be about 5 to 10 MHz when driving the external stray capacitance since it serves only as a buffer amplifier. That is, the amplifier 51 need not provide any actual signal gain and must only be capable of accurately following its input signal.
This completes the description of the preferred embodiments of the invention. Although preferred embodiments have been described, it is believed that numerous modifications and alterations thereto would be apparent to one of ordinary skill in the art without departing from the spirit and scope of the invention.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US4011442 *||Dec 22, 1975||Mar 8, 1977||General Electric Company||Apparatus for sensing optical signals|
|US4335406 *||Jun 26, 1980||Jun 15, 1982||Hitachi, Ltd.||Solid-state imaging device|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US4797561 *||Aug 28, 1986||Jan 10, 1989||Kyocera Corporation||Reading apparatus with improved performance|
|US4843473 *||Mar 21, 1988||Jun 27, 1989||Polaroid Corporation||Charge injection device with low noise readout|
|US5311319 *||Mar 27, 1992||May 10, 1994||Kabushiki Kaisha Toshiba||Solid state image pickup device having feedback voltage to amplifier|
|US5600451 *||Aug 10, 1994||Feb 4, 1997||Sony Corporation||Charge transfer device and output circuit thereof|
|US5821998 *||Aug 31, 1995||Oct 13, 1998||Nec Corporation||Method and apparatus for producing an image signal with an improved S/N ratio|
|US6239839||Feb 11, 1998||May 29, 2001||Kabushiki Kaisha Toshiba||MOS-type solid-state imaging apparatus using a unit cell formed of a photoelectric converter, amplification transistor, address capacitor, and reset transistor|
|US6538695||Nov 4, 1998||Mar 25, 2003||Ic Media Corporation||On-chip fixed-pattern noise calibration for CMOS image sensors|
|US6795121 *||Dec 12, 2000||Sep 21, 2004||Kabushiki Kaisha Toshiba||MOS-type solid-state imaging apparatus|
|US6801253 *||Mar 9, 1998||Oct 5, 2004||Sony Corporation||Solid-state image sensor and method of driving same|
|US6940553 *||Jan 26, 1998||Sep 6, 2005||Nec Corporation||Solid-state camera including a charge coupled device|
|US7132638 *||Jun 10, 2004||Nov 7, 2006||Fairchild Imaging||CCD imaging array with improved charge sensing circuit|
|US7369169||Aug 12, 2004||May 6, 2008||Kabushiki Kaisha Toshiba||MOS-type solid-state imaging apparatus|
|US7460164 *||Jan 21, 2005||Dec 2, 2008||Canon Kabushiki Kaisha||Image capture device having amplification circuit for amplifying signal from photoelectric conversion portion|
|US7893978 *||Mar 26, 2008||Feb 22, 2011||Canon Kabushiki Kaisha||Image capture device having amplification circuit for amplifying signal from photoelectric conversion portion|
|US8462242 *||Jan 11, 2011||Jun 11, 2013||Canon Kabushiki Kaisha||Image capture device having amplification circuit for amplifying signal from photoelectric conversion portion|
|US20010013901 *||Mar 19, 2001||Aug 16, 2001||Yoshiyuki Matsunaga||MOS-type solid-state imaging apparatus|
|US20020149688 *||Dec 12, 2000||Oct 17, 2002||Yoshiyuki Matsunaga||MOS-type solid-state imaging apparatus|
|US20050012839 *||Aug 12, 2004||Jan 20, 2005||Yoshiyuki Matsunaga||MOS-type solid-state imaging apparatus|
|US20050168603 *||Jan 21, 2005||Aug 4, 2005||Canon Kabushiki Kaisha|
|US20050274873 *||Jun 10, 2004||Dec 15, 2005||Xinqiao Liu||CCD imaging array with improved charge sensing circuit|
|US20110102654 *||May 5, 2011||Canon Kabushiki Kaisha|
|WO2005123228A2 *||Jun 2, 2005||Dec 29, 2005||Fairchild Imaging||Ccd imaging array with improved charge sensing circuit|
|U.S. Classification||348/241, 348/E05.079|
|International Classification||H04N5/369, H04N5/365, H01L27/146|
|Sep 16, 1985||AS||Assignment|
Owner name: FUJI PHOTO FILM CO., LTD., 210, NAKANUMA, MINAMI A
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST.;ASSIGNOR:TISUE, GIL;REEL/FRAME:004453/0862
Effective date: 19831214
|Apr 14, 1989||FPAY||Fee payment|
Year of fee payment: 4
|Apr 21, 1993||FPAY||Fee payment|
Year of fee payment: 8
|Jun 3, 1997||FPAY||Fee payment|
Year of fee payment: 12